XFA - Avoid to have containers not pushed in the html

[calixteman]

• it aims to fix issue Large amount of text missing #13668.

[calixteman]

/botio test

[pdfjsbot]

From: Bot.io (Linux m4)

---

Received

Command cmd_test from @calixteman received. Current queue size: 0

Live output at: http://54.67.70.0:8877/f4d0d42da931a81/output.txt

[pdfjsbot]

From: Bot.io (Windows)

---

Received

Command cmd_test from @calixteman received. Current queue size: 0

Live output at: http://3.101.106.178:8877/643358ce3e45445/output.txt

[pdfjsbot]

From: Bot.io (Linux m4)

---

Failed

Full output at http://54.67.70.0:8877/f4d0d42da931a81/output.txt

Total script time: 16.80 mins

• Font tests: Passed
• Unit tests: Passed
• Integration Tests: FAILED
• Regression tests: FAILED

  errors: 1238
  different ref/snapshot: 10

Image differences available at: http://54.67.70.0:8877/f4d0d42da931a81/reftest-analyzer.html#web=eq.log

[pdfjsbot]

From: Bot.io (Windows)

---

Failed

Full output at http://3.101.106.178:8877/643358ce3e45445/output.txt

Total script time: 19.18 mins

• Font tests: Passed
• Unit tests: Passed
• Integration Tests: Passed
• Regression tests: FAILED

  errors: 1226
  different ref/snapshot: 6
  different first/second rendering: 1

Image differences available at: http://3.101.106.178:8877/643358ce3e45445/reftest-analyzer.html#web=eq.log

[calixteman]

/botio makeref

[pdfjsbot]

From: Bot.io (Linux m4)

---

Received

Command cmd_makeref from @calixteman received. Current queue size: 0

Live output at: http://54.67.70.0:8877/cd1581e7061c8ae/output.txt

[pdfjsbot]

From: Bot.io (Windows)

---

Received

Command cmd_makeref from @calixteman received. Current queue size: 1

Live output at: http://3.101.106.178:8877/45e2d491c5300ce/output.txt

[Snuffleupagus]

Hmm; given that both browsers timed-out on both of the bots during the tests above, see #13716 (comment) and #13716 (comment), how do we know that this doesn't cause a bunch of breakage/regressions!?

[calixteman]

My bad, I missed that... I checked locally that everything was good and it was.
Anyway, do you prefer I revert and then push again, test, ... ?

[marco-c]

Hmm; given that both browsers timed-out on both of the bots during the tests above, see #13716 (comment) and #13716 (comment), how do we know that this doesn't cause a bunch of breakage/regressions!?

Where can I see an indication that they timed out?

[pdfjsbot]

From: Bot.io (Linux m4)

---

Failed

Full output at http://54.67.70.0:8877/cd1581e7061c8ae/output.txt

Total script time: 13.72 mins

• Lint: Passed
• Make references: FAILED

[Snuffleupagus]

I checked locally that everything was good and it was.

Strange, for me this patch actually times-out locally as well.

Where can I see an indication that they timed out?

There's errors: 1238 respectively errors: 1226 printed in #13716 (comment) respectively #13716 (comment), and those kind of (general) errors should never appear when testing a working patch. After that, following the "Full output at..." links, the reference tests end with:

TEST-UNEXPECTED-FAIL | test failed firefox has not responded in 120s
TEST-UNEXPECTED-FAIL | test failed chrome has not responded in 120s

Edit: Another indication of something being wrong is the total run-time of the tests, since the bots finishing in under twenty minutes hasn't happened in years (given the total number of added test-cases).

[pdfjsbot]

From: Bot.io (Windows)

---

Failed

Full output at http://3.101.106.178:8877/45e2d491c5300ce/output.txt

Total script time: 16.10 mins

• Lint: Passed
• Make references: FAILED

[marco-c]

There's errors: 1238 respectively errors: 1226 printed in #13716 (comment) respectively #13716 (comment), and those kind of (general) errors should never appear when testing a working patch. After that, following the "Full output at..." links, the reference tests end with:

Could the bot surface the "has not responded" errors in the comment or the HTML output?

[Snuffleupagus]

@calixteman I've tested this patch locally, in the viewer (with enableXfa = true set), and xfa_bug1717668_2 causes the browser (or tab rather) to hang (with full CPU utilization of one core) and the document doesn't render.

---

Could the bot surface the "has not responded" errors in the comment or the HTML output?

But as mentioned this is already indirectly visible, via the errors: {number here} lines, since those should never appear and that's thus a clear indication to look at the full logs.

[calixteman]

The same for me locally... I was sure I tested (and I'm still sure) but obviously I didn't.